FROM POLICY TO PRACTICE A Process Evaluation of Specialized Substance Abuse Supervision in Nebraska

FROM POLICY TO PRACTICE A Process Evaluation of Specialized Substance Abuse Supervision in Nebraska A final report prepared by the Vera Institute of Justice for the Nebraska Community Corrections Council Reagan Daly, PhD Maggie Peck Vera Institute of Justice May 1, 2009

Executive Summary In 2000, after decades of enacting mandatory minimum sentences and limiting early releases from prison, Nebraska faced its second prison population crisis in a decade. State leaders recognized the need for a new approach to control prison growth, while at the same time maintaining public safety and holding offenders accountable. In 2003, the Legislature created a new state entity the Community Corrections Council which was charged with creating a community corrections infrastructure for Nebraska. To date, the Council has supported the development of a number of community corrections options for nonviolent offenders. Most significant among these efforts is Specialized Substance Abuse Supervision (SSAS), an evidence-based supervision program for prison-bound felony drug offenders and early-release parolees that combines treatment with services that address other types of needs. SSAS was piloted by the Office of Probation Administration (OPA) in five sites in 2006 with the explicit goal of expanding statewide to reduce growth in the state s prison population. To facilitate a data-driven approach for moving forward, the Vera Institute of Justice conducted a process evaluation of the five pilot SSAS sites to explore the fidelity of program implementation and to better understand factors that influenced program implementation. The goal of this evaluation is to provide feedback to the Council and OPA on both of these questions that can be used to further develop the statewide implementation model. This process evaluation is also an essential pre-cursor to a proper outcome evaluation of SSAS. Without a process evaluation, it will be difficult to determine if changes in outcomes such as recidivism are due to the program itself or the way it has been implemented. This report summarizes our findings and recommendations. In brief, our findings include the following: Participants were invested in the SSAS program and felt a strong sense of collaboration with other partners. Despite this support, participants felt that implementation could be enhanced in a number of ways. Analysis of administrative data suggests that selected eligibility criteria are being implemented consistently; however, many people involved in the administration of SSAS think these criteria should be revised in particular the criterion related to crime of conviction. Probation and parole staff expressed a desire for clear and direct communication of procedures, policies, practices, and performance measures. This includes communication between administration and field officers and among field officers and staff who work in different areas of the state. Probation staff have difficulty using the agency s electronic case management system and other data collection tools and expressed concerns about the accuracy of data in the system. Vera Institute of Justice ii

Services are widely available to meet most client needs, although mental health and transportation needs remain prevalent. Based on these findings, we recommend the following: Enhance the collaboration that already exists between administrators, field officers, and program staff by bringing invested parties together more frequently for exchange of ideas and discussion of best practices in supervision and service delivery. Outline policies, procedures, and performance expectations in writing for probation and parole officers. Explore options for expanding resources and services to meet additional client needs including mental health treatment, transportation services, and the formation of an alumni network for program graduates who seek further support. Develop training procedures and other informational materials for SSAS officers and judges to enhance their knowledge of SSAS. Revisit the eligibility criteria for SSAS and in particular, the conviction criterion to assess the impact on excluding individuals who have drug-related criminogenic needs and could benefit from the program. Explore options for making Probation s electronic case management system and other data collection instruments more user-friendly. Vera Institute of Justice iii

Acknowledgments We would like to thank the following individuals for their assistance in conducting this study: Tina Chiu Joan Dietrich Mike Dunkle Ellen Fabian-Brokofsky Abbi Leman Peggy McGarry Deb Minardi Julie Rogers Christine Scott-Hayward Robert Sims David Wegner Dan Wilhelm Michael Woodruff We would also like to thank all of the probation officers, parole officers, day-evening reporting center coordinators, judges, service providers, and SSAS clients who participated in the interviews and focus groups as a part of this study. Vera Institute of Justice iv

Table of Contents Introduction... 1 Background... 3 Methodology... 5 Administrative data analysis of SSAS probation clients... 8 Qualitative analysis on the implementation of SSAS... 14 Recommendations... 20 Conclusion... 22 List of Figures and Appendices Figure 1: Flowchart of SSAS supervision... 4 Figure 2: Interview and focus group participants... 6 Figure 3: SSAS probation clients by SSI scores... 9 Figure 4: Top 10 conviction charges among SSAS clients... 10 Figure 5: Top 10 DRC program visits in January 2009 by type of program received... 11 Figure 6: DRC service visits in January 2009 by type of service received... 11 Figure 7: SSAS probation clients by revocation disposition... 13 Appendix 1: Overview of SSAS supervision stages... 23 Appendix 2: Snapshot of SSAS probation clients... 24 Appendix 3: Preliminary outcomes among SSAS probation clients... 30 Vera Institute of Justice v

Introduction For nearly a decade, officials in Nebraska have been developing a community corrections infrastructure to alleviate persistent prison population pressures while preserving public safety. In 2003, the Legislature enacted the Community Corrections Act, which, among other things, established the Community Corrections Council (Council) as the entity responsible for overseeing the development and advancement of community corrections for the state. With the support of the Council, the Office of Probation Administration (OPA) has taken the lead in implementing several new community corrections programs. Most significant among these efforts is the Specialized Substance Abuse Supervision (SSAS) program, an evidence-based supervision program for prison-bound felony drug offenders and early-release parolees that seeks to break the relationship between substance abuse and crime through treatment, cognitivebehavioral treatment, and intensive monitoring by specially trained probation officers. SSAS was piloted in five sites serving eight counties in 2006 with the explicit goal of expanding statewide to reduce growth in the prison population. 1 To facilitate a data-driven approach for moving forward, the Community Corrections Council requested that the Vera Institute of Justice (Vera) conduct a process evaluation of the five pilot SSAS sites to explore the fidelity of program implementation and to better understand factors that influenced program implementation. The goal of this evaluation is to provide feedback to the Council and OPA on both of these questions that can be used to enhance the statewide implementation model. This process evaluation is also an essential pre-cursor to a proper outcome evaluation of SSAS. Without a process evaluation, it will be difficult to determine if changes in outcomes such as recidivism are due to the program itself or the way it has been implemented. The evaluation focuses on five main questions related to the implementation of SSAS: To what extent were procedures implemented as intended? Was implementation of these procedures consistent across sites? What have been the biggest successes and challenges of implementation? How do different agencies and partners work together to supervise and provide services to SSAS clients? What occurs at each of the different stages of SSAS supervision? We explored each of these questions as they relate to six stages of the program sentencing/referrals, supervision, treatment provision, service provision at day-evening reporting centers (DRCs), violations/sanctions, and program completion/transition. 1 These counties include Douglas, Lancaster, Sarpy, Cass, Otoe, Buffalo, Dawson, and Dakota. Cass and Otoe Counties each have their own reporting center for SSAS services but are considered to be one site because funding is split between the two. The same situation exists for Buffalo and Dawson Counties. Vera Institute of Justice 1

This report begins with an overview of the SSAS program, followed by a brief discussion of the methodologies used in this evaluation. The next two sections present findings from quantitative and qualitative analyses, and the report concludes with recommendations for continuation and statewide implementation. Vera Institute of Justice 2

Background: Specialized Substance Abuse Supervision Specialized Substance Abuse Supervision (SSAS) is an evidence-based supervision program for otherwise prison-bound felony drug or DUI offenders and early-release parolees. It combines drug treatment with services that address other types of needs (e.g. educational, vocational and behavioral). On the front end, people are sentenced to SSAS as part of their probation supervision, while parolees are given the option of enrolling in SSAS as a condition of early release. The SSAS referral process has two steps. First, cases are screened pre-sentence for basic SSAS eligibility by a referring probation officer. If a person meets the eligibility criteria, he or she is then screened for suitability by a SSAS probation officer. To be considered suitable, a person must be in a contemplative state of change (i.e., the idea that change has been triggered) in one or more criminogenic needs, show antisocial patterns, have a need for drug or alcohol treatment, have no diagnosis of mental illness, or if mentally ill, be in appropriate treatment, and be serving at least an 18-month probation term. 2 Once the two-step screen has been completed, the results are submitted as part of the pre-sentence investigation (PSI) report for the judge s consideration at sentencing. Parole cases go through a similar eligibility and suitability screening process; referring parole officers conduct a basic eligibility screen, and then they conduct the suitability screen. SSAS parolees and probationers are supervised by SSAS officers, who manage cases and plan treatment programs. Parolees continue to have a parole officer who oversees his or her case in coordination with the SSAS supervisor. Services are delivered at DRCs with cognitive behavioral treatment as a core component along with vocational, educational, and drug or alcohol treatment services. DRC coordinators manage the provision of services and work collaboratively with SSAS officers to supervise SSAS parolees and probationers. Clients stay in the program until they meet completion requirements, which include a 22- week cognitive-behavioral program and substance abuse treatment and the establishment of a community support system. Once completion requirements are met, individuals serve the rest of their sentences on traditional or intensive supervision. Clients who fail to meet the conditions of supervision (e.g., do not comply with their treatment plan) face graduated sanctions in response to violations (as opposed to new convictions) and the potential return to prison. Figure 1 presents a flowchart of the SSAS supervision stages. For more detailed information on the components of each stage, refer to Appendix 1. 2 Criminogenic needs are factors underlying criminal behavior that can be addressed and/or improved through targeted programming. One example is substance abuse; another is antisocial attitudes. Vera Institute of Justice 3

Eligibility screening Suitability screening Sentencing/ enrollment Figure 1: SSAS supervision stages Treatment with registered provider Supervision: Regular drug testing Small caseloads DRC services: CBT Educational Vocational Life skills Graduated sanctions Completion: Cognitive classes Treatment Support system Transition to traditional or ISP probation Vera Institute of Justice 4

Methodology The methodology for this study included both quantitative and qualitative techniques, as well as a review of SSAS policies and procedures. We employed this approach to explore both processes and outputs related to implementation and to ensure that the specific areas of exploration were directly informed by the elements of supervision outlined in the Office of Probation Administration s (OPA) policies. Policy review Vera staff first built an understanding of how SSAS was intended to function through a thorough review of all policies, procedures, and instruments related to the supervision program. OPA provided information on screening procedures, relapse policies, case management components, and caseload standards, in addition to copies of screening instruments and administrative forms. This information was then used to develop research questions about specific elements of the SSAS program. Administrative data analysis To explore the fidelity of program implementation, we analyzed administrative data on SSAS probation clients. 3 Individual-level information was extracted from OPA s case management system, the Nebraska Probation Information Management System (NPMIS). NPMIS contains data on all probation cases and includes information on demographics, crime of conviction, assessments, drug testing, violations, and sanctions, among other things. The individual-level analysis is based on all individuals on SSAS supervision on or after October 1, 2007, a total of 352 individuals. 4 To explore the frequency with which different services and programs are used by probationers and parolees, we also analyzed aggregate data from the DRCs. Because DRC reporting forms were introduced in late 2008, we were only able to analyze data from January 2009. Finally, we analyzed aggregate level data on felony drug offender sentences from 2003 to 2009 to examine the extent to which SSAS is diverting a prison-bound population of offenders as opposed to net-widening by moving offenders who would otherwise be on probation or other community-based alternatives into more intensive supervision. Vera staff explored how different elements of the sentencing and supervision process are practiced on the ground in a descriptive analysis. Our analysis focused on three areas of implementation: sentencing and eligibility criteria, with an emphasis on whether or not the program is diverting otherwise prison-bound offenders; DRC services and programs; and sanctions and revocations. Because this analysis is based only on a sample of individuals on 3 We had also planned on using administrative data from the Parole Administration, but ultimately decided against it for two reasons: 1) Parole s database is more limited than Nebraska Probation Information Management System and would not provide us with much comparable information on parole clients in SSAS. 2) The number of parole clients is so small that a separate analysis on them would yield very little useful information at the aggregate level. 4 Data prior to October 1, 2007, were excluded due to concerns about quality. Vera Institute of Justice 5

SSAS beginning October 1, 2007, the patterns are not necessarily generalizable to all SSAS clients; however, given reasonably accurate data, they are representative of SSAS clients and processes during the study period. Qualitative analysis Because this evaluation focuses on the implementation process, we relied most heavily on qualitative data collected in interviews and focus groups with individuals who deliver and receive SSAS supervision and services. In these interviews and focus groups, we asked questions about how different elements of SSAS are practiced across sites, the nature of collaboration between involved parties, the elements of the program that have been easiest to implement, and the biggest successes and challenges that exist in following SSAS procedures and policies, among other things. Data for the qualitative analysis were collected during two trips to Nebraska in December 2008 and January 2009 and through phone interviews conducted between January and March 2009. Vera staff convened focus groups with referring officers, SSAS officers, parole officers, DRC coordinators, and service providers. We conducted one-on-one interviews with a sample of district and drug court judges in Lincoln and Omaha. We also conducted phone interviews with an additional sample of judges, a sample of SSAS clients, and a group of DRC coordinators who were unable to participate in the in-person focus group. 5 Each focus group contained up to seven participants. A summary of participants by group appears in Figure 2. Figure 2: Interview and focus group participants Group Type of interview Number of participants Judges One-on-one, in person and on phone 9 SSAS officers 2 focus groups 9 total DRC coordinators 2 focus groups (one by phone) 4 Parole officers 1 focus group 6 Referring officers (Probation and Parole) 1 focus group 1 probation officer 4 parole officers Service providers 1 focus group 4 5 Originally, Vera staff planned to conduct in-person interviews with SSAS clients during the second trip to Nebraska. We were unable to do so, however, because we did not yet have access to the list of SSAS clients at the time of the trip. As an alternative, we conducted phone interviews with selected clients upon return to New York. Originally, we planned to interview treatment providers as well; however, after conducting the first set of focus groups with probation and parole officers it became clear that treatment services do not fall under the umbrella of SSAS policies and procedures in the same way as other elements of the program. While OPA has outlined expectations for contacts between providers and officers, the treatment approach is decided by the individual provider. Because of the largely independent role of treatment providers in the administration of SSAS, it seemed that any information they provided in interviews would be only minimally useful for informing Probation policies and procedures. As a result, a decision was made not to interview them. Vera Institute of Justice 6

SSAS clients One-on-one phone interviews by phone 9 (8 probation, 1 parole) Participants were selected using two different sampling approaches. For SSAS officers and DRC coordinators, we attempted to include as many individuals in the study as possible. Both of these roles are significant in the delivery of SSAS, and there were not enough people in either group to select a random sample (13 SSAS officers and 5 coordinators). Service providers, referring officers, parole officers, judges, and clients were selected randomly using a multi-stage sampling frame that allowed for representation from different SSAS sites, different Department of Correctional Services regions in the case of parole officers, and individuals on probation and parole. During focus groups and interviews, Vera staff tried to capture as much of the participants perspectives as possible through detailed note-taking and observation. We also recorded inperson interviews and focus groups with participants permission. Qualitative data were analyzed for content, and overarching themes were identified by synthesizing findings across interviews and focus groups. Vera staff paid particular attention to analyzing commonalities and differences in the successes and challenges of practicing SSAS across the five sites and supervising probation clients compared to parole clients. Vera Institute of Justice 7

Administrative data analysis of SSAS probation clients This section presents findings from the administrative data analysis. The purpose of this analysis was to explore the fidelity with which selected elements of the program were implemented across the five pilot sites, with an emphasis on three areas: 6 Sentencing and eligibility criteria, including whether or not SSAS is diverting prisonbound offenders Sanctions and revocations DRC programs and services These areas were chosen based on the measures available in NPMIS to test implementation. Results presented in this section are based on data elements believed to have the highest degree of accuracy; for data elements in which concerns were expressed, this is noted in the report. 7 Vera staff worked with data analysts in OPA in the collection, analysis, and interpretation of NPMIS data for this analysis. Broadly, the analysis revealed that some measurable eligibility criteria are being implemented with high fidelity across the clients in this sample. Consistent with SSAS policies, approximately three-quarters of the sample scored in the moderate-high range on the Simple Screening Instrument (SSI), and the large majority of clients were convicted of at least one felony-offense involving drugs or alcohol. Of the small proportion who were not convicted of a felony offense involving drugs or alcohol, most had either a misdemeanor conviction or an aggravating factor linked to substance abuse. With respect to diverting prison-bound offenders into SSAS as opposed to net widening, 8 aggregate data suggests that this is occurring, although more data is needed to make a sound conclusion about this question. The implementation of other program elements is more difficult to assess due to data limitations, but the data show that probationers and parolees visit DRCs most often for cognitivebehavioral treatment, GED classes, or drug testing all core components of SSAS. The analysis did not provide any definitive conclusions on the use of graduated sanctions. These findings are discussed in more detail in the remainder of this section. 6 We also analyzed background characteristics and some preliminary outcomes for the sample of SSAS clients. The findings of those analyses are presented in Appendices 2 and 3. 7 This analysis excludes any data determined to be so unreliable that it would not provide any meaningful information to readers. For example, OPA expressed concerns about the quality of program end dates, so we did not include any findings related to length of time in the SSAS program. We opted to leave other data in the analysis with qualifications, however, if we knew the data tended to be biased in one way or another. For example, OPA was concerned that officers underreport sanctions, so we left that analysis in the report, with the caveat that our numbers likely underrepresent the prevalence of sanctions among the study sample. 8 Net widening refers to failure of community corrections to divert the appropriate target population from prison, but rather provide a richer benefit program to already probation-bound clients. Vera Institute of Justice 8

Officers are using eligibility criteria related to substance use screening and crime of conviction with high fidelity among this sample. SSAS policies specify that the SSI, a substance abuse screening instrument, must be administered as part of eligibility screening, and individuals must score in the moderate to high range (four or higher) to be eligible for the program. Our analysis revealed that both of these policies were implemented consistently by the officers for this sample of clients. Only 8 percent of the sample was missing an SSI score, which suggests that officers are conducting these assessments as a regular part of the screening process. We also found that the majority of SSAS probation clients have moderate to high scores on the SSI. 9 As shown in Figure 3, almost 75 percent of the sample falls into this range. Figure 3: SSAS probation clients by SSI scores 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 74.1% 11.4% 6.5% 8.0% Low Minimal Moderate-high Missing Crime of conviction is among the most important criteria related to SSAS eligibility. The program is intended to target individuals convicted of felony drug or DUI offenses, and this is largely reflected in sentencing practices during the last year. Eighty-four percent of the sample had at least one felony substance abuse charge associated with their conviction. Of the 16 percent who did not, 5 percent were convicted of a misdemeanor substance abuse charge, and most of the remaining 11 percent had at least one aggravating factor linked to substance use or delivery. These trends are further reinforced upon examination of specific convictions among the sample. Figure 4 presents the ten most prevalent conviction charges associated with sentences to SSAS. 10 Nine of the ten charges are drug- or alcohol-related (the exception being criminal 9 SSI scores range from 0 to 14. A score of 0-1 indicates no to low risk; 2-3 minimal risk; 4 or more moderate to high risk 10 For individuals who had multiple charges associated with their convictions, we used the most serious conviction charge in this analysis. Most serious charge, in turn, was defined using offense classification (felony 1, felony 2, Vera Institute of Justice 9

attempt, 11 which accounts for 6.5 percent of the sample), and the most common conviction charge is possession of amphetamines, which is accounts for 24.7 percent of the individuals in the sample. Convictions for possession with intent to deliver and possession of a pharmaceutical controlled substance are also prevalent, accounting for 15.6 percent and 15.3 percent of the sample, respectively. These same patterns emerge when all conviction charges are analyzed. Figure 4: Top 10 conviction charges among SSAS probation clients Charge Number Percentage Amphetamine--possession 87 24.7 Possession with intent to deliver 55 15.6 Pharmaceutical controlled substance--possession 54 15.3 Criminal attempt 23 6.5 Amphetamine--possession with intent 21 6.0 Driving under the influence of liquor 20 5.7 Cocaine--possession 18 5.1 Marijuana--selling 13 3.7 Cocaine--possession--with intent 7 2.0 Marijuana--possession--more than 1 lb. 7 2.0 Since SSAS was piloted in 2006, sentences of felony drug offenders to DCS for three years or less have decreased by 69%, while sentences to SSAS probation increased 166%. This suggests that SSAS, in conjunction with other community corrections programming including drug courts, is targeting prison-bound individuals rather than traditional probationers. As previously discussed the target population for SSAS probation is individuals convicted of felony drug and DUI offenses 12 who have traditionally been sentenced to prison rather than traditional probation. Although no further criteria defining the target population within prison have been identified, sentencing guidelines that were proposed for felony drug offenses in 2007 specify that individuals with a minimum sentence of one year or less would be targeted for intermediate supervision, including SSAS. 13 This implies that the subpopulation of felony drug offenders in prison who would be most appropriate for diversion are those serving a sentence of three years or less. Figure 5 shows sentencing trends for felony drug offenders between 2003 and 2009 statewide, with sentences broken down into four categories: 1) Department of Correctional Services (DCS) on a sentence of more than three years; 2) DCS on a sentence of three years or less; 3) non-ssas Probation; and 4) SSAS probation 14. etc.). Finally, if an individual had multiple charges of the same classification, we prioritized substance use offenses over other types of offenses. 11 Many of the cases where criminal attempt was the most serious conviction charge did not have a second charge associated with it. 12 Felony DUI offenses became eligible for SSAS in July 2008. 13 The sentencing range would vary depending on the person s criminal history. 14 Admissions to SSAS broken down by probation and parole were only available from November 2007 onward. Because of this, we were not able to subtract those probationers from the total probation sentences between 2006 Vera Institute of Justice 10

Figure 5: Trends in sentencing for felony drug offenders, 2003-2009 Number sentenced 200 180 160 140 120 100 80 60 40 20 0 Q1 2003 Q3 2003 Q1 2004 Q3 2004 Q1 2005 Q3 2005 Q1 2006 Q3 2006 Q1 2007 Q3 2007 Q1 2008 Q3 2008 Q1 2009 DCS > 3yrs DCS 3 years or less SSAS probation Non-SSAS probation As shown, the number of felony drug offenders sentenced to DCS for three years or less decreased by 69% since SSAS was piloted in 2006, while DCS sentences for more than three years increased by 9% during the same time period. This suggests that SSAS is diverting the appropriate population of offenders from DCS, although without controlling for arrest trends and trends in sentences to drug courts and other community-based alternatives, 15 it is not possible to attribute this decline entirely to SSAS. 16 At the same time that DCS admissions were decreasing, there was also a sizable decrease in admissions to probation for felony drug offenses, which could be interpreted several ways this could indicate that SSAS was net widening by targeting some individuals who would have been sentenced to probation anyway; it could be reflective of statewide trends in probation sentences rather than those in SSAS counties; or it could be reflective of a decrease in the number of felony drug arrests during that time period rather than a change in sentencing practices for those individuals. Because the data do not allow us to explore these explanations, the strongest conclusion that can be drawn from this analysis is that these trends are not inconsistent with the idea that SSAS is diverting prison-bound individuals. Consistent with the identified core components of SSAS, the most prevalent programming and service visits to DRCs are for moral reconation therapy (MRT), GED classes, and drug testing. and November 2007, and trends in non-ssas probation sentences during that time are reflective of both SSAS and non-ssas probationers. 15 Admissions to drug court were only available for 2007 and 2008, and were not available on a quarterly basis. 16 It is also important to consider that SSAS was only piloted in eight counties, and these trends represent sentencing patterns at the state level. Because of this, we do not know how DCS sentences have changed in those counties as a result of SSAS. Vera Institute of Justice 11

Aggregate-level data on day-reporting center contacts were used to explore the frequency at which different services and programs were utilized. Because the reporting system for DRCs is relatively new, only data for the month of January 2009 could be obtained. As a result, it was not possible to identify trends in service utilization over time. In total, there were 1,555 visits to DRCs by SSAS clients in January 2009: 261 to participate in programming that addresses particular needs, and an additional 1,294 visits to participate in services that do not directly address underlying criminogenic needs, such as drug testing or child care. 17 Of greater relevance is a breakdown of visits by type of service or program received, which is depicted in Figures 5 and 6 (separate charts are presented for services and programs). Figure 5 presents the ten most prevalent programming visits during the month of January. Moral reconation therapy (MRT), a form of cognitive-behavioral treatment, accounts for the largest proportion of visits (41.4 percent), followed by GED classes (11.9 percent), vocational services (through Goodwill) (5.4 percent), and relapse prevention (5.4 percent). When we examine service visits, we see that they are overwhelmingly comprised of drug tests (98.8 percent of the total). These findings are consistent with SSAS requirements specifying frequent drug testing, MRT, and educational services as core components of service provision. Figure 6: Top 10 DRC program visits in January 2009 by type of program received 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% MRT 41.4% 11.9% GED Vocational 5.4% 5.4% 5.4% 5.0% 4.6% Relapse Prevention Voc. Rehab Pre-treatment Family Orientation SSAS Budgeting/Money Skills 3.4% 2.7% 2.3% T4C AA / Alanon N=261 visits for programming 17 We did not have information on the number of people who visited during this time period. These visits were broken down by DRC as well, although the numbers are not of great significance without knowing the number of clients who reported to each center during the time period as well. Vera Institute of Justice 12

Figure 7: DRC service visits in January 2009 by type of service received 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Transportation 98.8% 0.7% 0.0% 0.5% 0.0% 0.0% Drug Testing N=1,294 visits for services Child Care Computer Lab Mediation Vocational grad The use of graduated sanctions is unclear. SSAS procedures specify that graduated sanctions should be employed for an individual until they have been exhausted, so we expected to find a substantial percentage of clients who had been sanctioned during the course of supervision. However, the large majority of individuals who were on SSAS on or after October 1, 2007 (73 percent) did not receive any sanctions, and an even larger majority did not receive any sanctions for substance abuse violations (75.9 percent). The low prevalence of sanction events does not necessarily indicate that officers are not employing graduated responses to violation behaviors though. It could be that officers are underreporting sanctions, a concern that has been expressed by OPA. Alternatively, the low prevalence could reflect a low prevalence of violation behaviors or be a function of people not being on the program for a long enough period of time. Unfortunately, we cannot test any of these explanations because we do not have data on violation behaviors that are not associated with sanctions, nor do we have access to an alternative data source on sanction activities or enrollment dates. The data do show, however, that of clients who were sanctioned during the study period, the majority had at least one substance abuse violation (89.5 percent). The data also indicate that only a small proportion of the sample (10.8 percent) was revoked during the study period, and that three-quarters of revocations were associated with law violations (as opposed to technical violations). This proportion is clearly smaller than the Vera Institute of Justice 13

proportion sanctioned (a difference of roughly 17 percentage points), which suggests that officers are employing graduated responses to some extent before they request a revocation warrant. Furthermore, the average number of sanctions administered to clients who were revoked is higher than the average for clients who were not revoked (1.58 vs. 0.34, respectively). Because the latter average includes a substantial number of people with no sanctions, however, it is difficult to know whether this trend reflects the use of graduated sanctions, the fact that most people who were not revoked did not engage in any behavior that would warrant a sanction or revocation, or the fact that most people in the sample were not sanctioned. 18 Figure 7 presents findings on the dispositions following revocation events, and although a large number are still pending (N=8 out of 38), we can see that roughly one quarter of revoked clients were sentenced to prison or jail. A noticeable number were returned to probation as well (eight out of 38). Figure 8: SSAS probation clients by revocation disposition 18 When people with no sanctions are excluded from the analysis, the numbers are much more similar (2.00 vs. 1.66, respectively), which suggests that people with no sanctions are driving the initial difference. The average remains higher among people who were revoked. Vera Institute of Justice 14

Number of Revocations 9 8 7 6 5 4 3 2 1 0 5 2 8 5 5 3 3 7 Dept of Corrections Intensive Supervsn Jail Pending Probation Continued Probation Extended Other Missing N=38 Other category includes Work Ethic Camp, SSAS intensive supervision, and dismissed cases. Vera Institute of Justice 15

Qualitative analysis on the implementation of SSAS As discussed earlier, the purpose of the qualitative analysis was to better understand factors and processes related to the quality of SSAS implementation, including both the successes and challenges of this process, the nature of collaboration between partners, and the extent to which supervision practices and other program components vary across sites. This section presents findings related to these issues, drawing on perspectives from SSAS officers and parole officers, DRC coordinators, judges, service providers, and SSAS clients. Broadly, our interviews and focus groups led to findings in four main areas: General Perspectives on SSAS The biggest success of SSAS implementation to date is the widespread support for the program. Across the board, participants were invested in the SSAS program and felt a strong sense of collaboration with other partners in the initiative. At the same time, participants felt that implementation could be enhanced in a number of ways. Clients were responsive to SSAS and found the program helpful. Communication and Training Probation and parole staff felt that one of the most challenging aspects of supervising SSAS clients is their lack of clarity on procedures, policies, practices, and performance measures. They expressed a desire for clear and direct communication on all of these. This includes communication between administration and field officers and among field officers and staff who work in different areas of the state. There is a great deal of inconsistency across sites, both in terms of how services and supervision are delivered, the challenges that exist, and the populations served. Resources and Service Delivery In general, services are widely available to meet the needs of SSAS clients, but mental health and transportation services remain prevalent needs. SSAS Policies and Procedures People involved in the administration of SSAS think that eligibility criteria should be revised. Of particular concern are the criteria related to crime of conviction and, in the case of parolees, restrictions related to completion of drug treatment in prison. There are practical challenges involved in the supervision of parole clients on SSAS, ranging from communication to supervision logistics. Participants felt that many of these challenges could be addressed through the development of specialized procedures for the supervision of parole clients on SSAS. Vera Institute of Justice 16

Probation staff have a difficult time using NPMIS and other data collection tools and expressed concerns about the accuracy of data in the system. These findings are discussed in more detail in the remainder of this section. General perspectives on SSAS In general, interviews and focus groups with study participants revealed considerable support for the SSAS program. Across the different groups, participants felt that SSAS was an important initiative that addressed the gap in programmatic and treatment options available to offenders in the community. SSAS officers pointed out that little had been done to change OPA for many years and that they welcome the philosophical change in offender management. Participants across the board were pleased with the collaborative efforts between OPA and the Parole Administration, and most felt that they were on the same page as other partners. Some even said that strong collaboration between partners, ranging from referring officers to treatment providers, was one of the biggest advantages of the program. When asked for their views on the purpose of SSAS, most study participants spoke about the rehabilitative impact on individuals rather than its impact on the prison population. The one exception was parole officers, who identified themselves as more law enforcement-oriented and said that they have less tolerance for relapse and other non-compliant behavior among the people they supervise. One judge also questioned the cost-effectiveness of having a separate program for drug offenders rather than more services for traditional probationers, but like the majority of study participants, this judge s sentencing philosophy was closely aligned with the rehabilitative ideal. Finally, our interviews with clients suggest that they also support the program, or at least those under supervision who referred to themselves as motivated to change. This was the case for most of the individuals who were interviewed for this study, and they said that SSAS provides the structure and support necessary to help them change. 19 A number of participants mentioned the relationship with their SSAS officer in particular as an integral component of their success in the program. This is consistent with officers perspectives on SSAS clients. In general, they felt that their relationships with clients were good and that clients benefited from the program. A few officers even mentioned that some of their supervisees have intentionally violated supervision conditions to keep from transitioning out of the program. Communication and training One of the primary themes that emerged from interviews and focus groups was a desire among probation and parole staff for clear and direct communication, both from the administration regarding policies and procedures related to SSAS and among sites regarding supervision 19 Because we only spoke to a small sample of individuals from the program, these findings are not necessarily representative of all clients on SSAS. Vera Institute of Justice 17

practices. With respect to the communication of policies and procedures, one of the concerns expressed by SSAS officers was a lack of clarity about how well they are doing their jobs. To address this, a number of them suggested the inclusion of performance measures in policies. Officers also felt that new policies and procedures could be better communicated to line staff. This was a particularly salient issue for parole officers, who felt not only that they did not have a clear sense of the SSAS policies, but that changes in policies were not communicated to them effectively. There was even an instance of this during the course of the study. In January 2009, screening procedures changed so that parole officers became responsible for suitability screenings. Based on discussions that took place in the focus groups, however, it did not appear that this change was communicated to parole officers before it took effect. Similar to probation and parole officers, judges expressed a desire for more frequent communication from the administration as well. Several of the judges with whom we spoke had a limited knowledge about the SSAS program and general programming options for defendants. For example, one judge said that he had little understanding of the types of services available to defendants at the DRCs; another judge said that he did not know the difference between SSAS and drug court. It should be noted that this was limited to non-drug court judges; district court judges that also preside in drug court had a much better understanding of the programmatic options available to defendants in the community. In addition to improved communication of policies and procedures from the administration, some of the participants also expressed a desire for more communication across sites. As discussed earlier, one of the main questions explored in this process evaluation was the consistency of SSAS supervision across sites. During our focus groups with probation staff, it became apparent that not only are there many differences in how SSAS is practiced across sites, but also that officers found it helpful to come together and hear about what others are doing. Among the differences that emerged in our focus groups were practices related to service provision, supervision structure, and program and completion requirements. For example, officers in different sites have different processes for transitioning clients onto regular probation supervision. Many, but not all, use a phase or step-down system to transition clients, but the number of phases varies across sites. Cognitive-behavioral treatment is incorporated into supervision differently across sites. Some counties do not require participation in cognitivebehavioral classes if clients completed it in prison, while others require it regardless of previous experience. Based on these discussions DRC coordinators and SSAS officers expressed a desire for more frequent and consistent communication across sites to share information and best practices. Finally, both probation and parole officers felt that improvements in training would be a good way to address these challenges related to understanding SSAS procedures. Many of the officers said that they did not have a sense of how the program was supposed to run after participating in the initial six-week SSAS training because the training provided a general overview of evidence-based practices but did not go into detail on the specifics of the SSAS program components. Furthermore, since the program was originally piloted, the duration of Vera Institute of Justice 18

training for new officers has been reduced from six weeks to one week. While some participants thought that one week was enough, a number said that longer trainings would be beneficial, at least for some of the new officers. One participant in particular felt that it would be helpful for new SSAS officers who were previously traditional probation officers, given that they are generally less accustomed to a rehabilitative supervision approach. Resources and service delivery Interviews and focus groups also revealed the existence of a continuum of services available to SSAS clients. Across the different groups, participants felt that services designed to meet the needs of SSAS clients were widely available throughout the selected pilot counties and especially in urban areas. Furthermore, DRC coordinators work with their SSAS officers to assess and update service provision on a regular basis. According to coordinators, SSAS officers are instrumental in providing information on additional services needed by clients, recommendations about particular service providers, and feedback on the day-to-day functioning of the centers. This allows coordinators to adapt service provision to best meet the needs of clients they serve. Despite the range of services available, addressing co-occurring disorders (i.e. combined mental health and substance abuse issues) and transportation needs continues to be a challenge. Probation officers and DRC coordinators said that co-occurring disorders are prevalent among the people they supervise, and officers felt strongly that clients should be given a mental health screen in addition to a substance abuse screen. Many of them pointed out though that even if they had the capacity to do such screens they still would not have access to resources for people who were identified as having mental health problems. It seems that only the Lancaster reporting center has the ability to both screen and refer people to psychological counseling. Service providers who participated in the study suggested a voucher program to address these issues, similar to that of the substance abuse evaluation and treatment vouchers put into place by OPA in 2006. Transportation is a challenge as well, particularly for the Sarpy and Lancaster sites where clients may have to travel up to 25 miles to attend programming. Similar to their thoughts on mental health needs, both service providers and SSAS officers were supportive of a transportation voucher program for SSAS clients whose options are most limited in this area. Specific SSAS policies and procedures Finally, while not directly related to implementation, the following findings specific to SSAS policies and procedures are worth directed attention. Our interviews and focus groups revealed that probation and parole staff have concerns and/or difficulties related to specific procedures and policies. One of the most prevalent was a concern with the current eligibility criteria for the program. As described in the background section, to be recommended for SSAS, an individual must meet a number of requirements, including being convicted of a felony drug or DUI offense, Vera Institute of Justice 19

being high to very high risk in antisocial patterns and drug/alcohol patterns on the Level of Service/Case Management Inventory risk/needs assessment, and being in a contemplative state of change (i.e., considering change) in one or more areas of criminogenic thinking. Individuals can also be excluded for certain reasons, for example, if they have a mental illness that is not being treated or if their supervision sentence is too short. Across the groups we interviewed, there was a consistent concern with the conviction criterion in particular. Participants felt that limiting program eligibility to only individuals with felony drug or DUI convictions excludes a significant population of people who have criminogenic substance abuse needs and could benefit from SSAS, including people who plead down to misdemeanor drug charges and people convicted on non-drug charges (such as theft or burglary) whose criminal behavior is associated with substance use. Participants on the whole felt that SSAS should be available to anyone who has a drug problem. There was a more mixed response to the inclusion of felony DUI offenses, which became eligible during July 2008. A substantial number of participants were against it, although their rationale varied. According to probation officers, DUI offenders do not exhibit the same criminogenic thinking patterns as drug offenders and therefore should be supervised in separate units so as not to dilute the integrity of the programming. This is already being done in Lincoln, where there is a large enough population of SSAS clients to fill all of the SSAS slots with felony drug offenses. It is an issue in smaller counties, though, where they do not have the same numbers of felony offenses. Judges had a more mixed response to the inclusion of DUI cases. Some felt it was a good idea as long as they would benefit from the program; others thought that felony DUI cases pose too great a threat to public safety to be supervised in the community. Despite some disagreement on who is appropriate for the program, participants agreed that SSAS should not be a numbers game that focuses too much on filling slots. Our interviews and focus groups also revealed eligibility concerns specific to parole clients, although again, these perspectives were mixed. Both referring parole officers and field parole officers identified the exclusion of people on parole who completed drug treatment in prison as problematic. Reentry parole officers, who conduct eligibility screenings in prison pre-release, felt that people who participate in institutional drug treatment should still be eligible for SSAS, namely because staying drug-free in the community is much more difficult than doing it in an institution where there are fewer temptations and opportunities for relapse. Field parole officers, on the other hand, were supportive of this criterion. The group of officers who participated in this study said that their clients are dealing with too many other issues upon return to the community to focus on drug treatment. They also thought that people should be given a chance to fail before they are put into a treatment program and that clients are most receptive to SSAS after a violation when they are faced with the possibility of returning to prison. They liked having SSAS as an option for responding to violations. A second theme that emerged in the area of policies and procedures was the challenge of supervising parole clients on SSAS. Both probation and parole officers identified a number of obstacles associated with this particular duty. First, although parole officers remain the primary Vera Institute of Justice 20